Awesome

BEUrRE

Resources and code* for paper "Probabilistic Box Embeddings for Uncertain Knowledge Graph Reasoning"

Install

Make sure your local environment has the following installed:

Python3.7
torch == 1.4.0
wandb == 0.9.7

Install the dependents using:

pip install -r requirements.txt

Run the experiments

To run the experiments, use:

python ./main.py --data cn15k --task mse

• You can switch to NL27k using --data nl27k
• To train and test for the ranking task, use --task ndcg

To test a trained model, you can use the following command:

python ./test.py --data nl27k --task mse --model_path ./beurre-pretrained-models/nl27k-mse.pt

The pre-trained models are available here here.

Reference

Please refer to our paper.

Xuelu Chen*, Michael Boratko*, Muhao Chen, Shib Sankar Dasgupta, Xiang Lorraine Li, Andrew McCallum. Probabilistic Box Embeddings for Uncertain Knowledge Graph Reasoning. Proceedings of the 19th Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), 2021

* Indicating equal contribution

@inproceedings{chen2021boxukg,
    title={Probabilistic Box Embeddings for Uncertain Knowledge Graph Reasoning},
    author={Chen, Xuelu and Boratko, Michael and Chen, Muhao and Dasgupta, Shib Sankar and Li, Xiang Lorraine and McCallum, Andrew},
    booktitle={Proceedings of the 19th Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL)},
    year={2021}
}